---
language:
- en
license: cc-by-nc-4.0
pipeline_tag: image-text-to-text
tags:
  - multimodal
base_model:
  - Qwen/Qwen2-VL-7B-Instruct
---
# S1-M-7B-Beta

[🏠 Homepage](https://github.com/PKU-Alignment/s1-m) | [👍 Our Official Code Repo](https://github.com/PKU-Alignment/s1-m) | [🤗 S1-M Dataset (Beta)](https://huggingface.co/datasets/PKU-Alignment/s1-m_beta) 

S1-M-7B-Beta used for developing the algorithm "Simple Test-time Scaling in Multimodal Reasoning". By fine-tuning the base model `Qwen/Qwen2-VL-7B-Instruct` on data with thinking tags `<think>` and `</think>`, the model acquired the `think first, then response` paradigm, allowing for experiments on "Test-time Scaling".

**Note: The current model is a development version, not the final official version.**